I have seen that problem too.
For DIY this is not a serious problem, since you
can (hopefully) guarantee, that the speaker is
moved by "trained" personnel only
@GM: OK, historical solutions for PA installations acknowledged.
One could use a transport lock like in a washing machine ...
Has anyone used the transport lock of his washing machine
the last time it moved ??? Or the transport lock of his CD player ???
OK. Doesn't work ...
Leads to some requirements, which had to be met in former designs too:
- The suspension has to be rigid enough to tolerate beeing moved
around and stay for some time in a "wrong" orientation in space
without deformation. (Define "some time").
- That implies "fd" the resonance of the (whole) driver & suspension
not beeing too low, because the suspension has to be stiff enough
for a given mass of driver to make only small excursion when a static
force (weight) is applied.
- The reset force should come from a material which is a real "spring"
operating always in its linear range.
- Damping comes from a mechanism/material which should be durable,
since with loss of damping system will get out of control when
excited with low frequencies.
Some proposals for the suspension:
Spring & Damper
---------------
Rubber/Elastomer & Rubber/Elastomer (spring and damper is the same)
Metal & Dashpot (air friction only, has to be manufactured with good precision)
Metal & Rubber/Elastomer
Last edited:
Used to decouple the tweeter assembly on the PEARL PR-2.
The designer is a new member, maybe he will chime in.
dave
... buying a forklift instead of a new car can result in loud noises not radiating from the speaker.
I must admit that non speaker born sound can be an issue too ...
You guys are overthinking this. Soft foam will give compliance but you can still have centering pins in holes to fix location as long as hard objects don't touch hard objects.
The cheaper KEF units used foam tape behind the chassis and screws through rubber grommets (just like chassis/wire grommets) to hold the chassis to the cabinet and define location. A metal sleeve in the center of the grommet prevented the screw from crushing it.
The larger units used proper vibration mounts.
See the motor issolation schemes in any old turntable with a large motor. Picture that to hold the driver in position and foam to seal the necessary gap between the chassis and baffle.
David
The cheaper KEF units used foam tape behind the chassis and screws through rubber grommets (just like chassis/wire grommets) to hold the chassis to the cabinet and define location. A metal sleeve in the center of the grommet prevented the screw from crushing it.
The larger units used proper vibration mounts.
See the motor issolation schemes in any old turntable with a large motor. Picture that to hold the driver in position and foam to seal the necessary gap between the chassis and baffle.
David
A mechanically isolated tweeter head
Here's a shot of the spring isolated tweeter head assy. (hopefully, this is the first time I've done this) from the Perkins PR 2, circa 1985. The springs are underneath and the backed out, large headed screws you see are what keep in it place and, screwed down, secure the assy. for transit.
Bill Perkins - PEARL
Here's a shot of the spring isolated tweeter head assy. (hopefully, this is the first time I've done this) from the Perkins PR 2, circa 1985. The springs are underneath and the backed out, large headed screws you see are what keep in it place and, screwed down, secure the assy. for transit.
Bill Perkins - PEARL
Attachments
I'm not sure what issolation mounts under the tweeter will achieve. It would be very hard to demonstrate that any woofer cabinet motion would generate significant intermodulation if transferred to the tweeter.
The first question is what are we trying to achieve?
We don't care if the woofer chassis moves a little or if the tweeter moves a little. The loss of output from the woofer because the chassis moves backwards slightly as the cone moves forward (remember the ratio of the masses) is insignificant and unmeasurable. What is nasty, is a high Q cabinet resonance coupled to a surface several times larger than the woofer cone.
If you look up a text book on architectural acoustics you will see measurements of wall transmission loss (TL) and modes called coincidence frequencies were the massive wall is virtually transparent. Same thing with speaker cabinets.
David
The first question is what are we trying to achieve?
We don't care if the woofer chassis moves a little or if the tweeter moves a little. The loss of output from the woofer because the chassis moves backwards slightly as the cone moves forward (remember the ratio of the masses) is insignificant and unmeasurable. What is nasty, is a high Q cabinet resonance coupled to a surface several times larger than the woofer cone.
If you look up a text book on architectural acoustics you will see measurements of wall transmission loss (TL) and modes called coincidence frequencies were the massive wall is virtually transparent. Same thing with speaker cabinets.
David
My first question to you is: Have you actually done the work or are you giving air to mere speculation ? ? I'd appreciate a simple 'Yes' or 'No' answer to the part of the question having to do your actually having done the work.
You allege, and wrongly, "We don't care if the woofer chassis moves a little or if the tweeter moves a little."
You might not but I care a great deal and do so the basis of many, many years, several decades in fact, of 'down in the trenches' loudspeaker design work with a half million dollar lab of gear that these days includes a $200,000.00 Polytec Scanning Laser Vibrometer; always a full-on woodshop and a very complete machine shop that days includes a CNC lathe and milling machine that I am -slowly- bringing on-line.
Presuming my ignorance, you go on to say, "If you look up a text book on architectural acoustics you will see measurements of wall transmission loss (TL) and modes called coincidence frequencies were the massive wall is virtually transparent. Same thing with speaker cabinets."
Please, I am perfectly well aware of cabinet resonance issues and successfully engaged that difficult matter in about 1984/5, doing so after about 5 continuous years work on the problem and well before the likes of B & W and Wilson Audio put forward their solutions. This, after much work with the BBC's then notion of thin-walled, unbraced cabinets with bitumen damping panels; sand filled panels; heavy tar-based damping methods; massive asymmetrical bracing; mass loaded, decoupled drive unit mounting; tuned anti-resonant masses; tar/sand damped vitreous structures; and so on.
So perhaps it's better to ask some questions than posit insubstantial hypotheses alongside assumptions as to another's technical acumen, particularly when said 'other' is one you probably do not even know.
Coming off looking as though you presume to teach your old grannie how to suck eggs will never put you in good stead.
I apologize if my tone comes across a little harsh, I've been in audio-related forums for 15 years, written probably thousands of posts and have by now seen a great deal of people's assumptions and very, very little of their simply asking questions.
Bill Perkins - PEARL
You allege, and wrongly, "We don't care if the woofer chassis moves a little or if the tweeter moves a little."
You might not but I care a great deal and do so the basis of many, many years, several decades in fact, of 'down in the trenches' loudspeaker design work with a half million dollar lab of gear that these days includes a $200,000.00 Polytec Scanning Laser Vibrometer; always a full-on woodshop and a very complete machine shop that days includes a CNC lathe and milling machine that I am -slowly- bringing on-line.
Presuming my ignorance, you go on to say, "If you look up a text book on architectural acoustics you will see measurements of wall transmission loss (TL) and modes called coincidence frequencies were the massive wall is virtually transparent. Same thing with speaker cabinets."
Please, I am perfectly well aware of cabinet resonance issues and successfully engaged that difficult matter in about 1984/5, doing so after about 5 continuous years work on the problem and well before the likes of B & W and Wilson Audio put forward their solutions. This, after much work with the BBC's then notion of thin-walled, unbraced cabinets with bitumen damping panels; sand filled panels; heavy tar-based damping methods; massive asymmetrical bracing; mass loaded, decoupled drive unit mounting; tuned anti-resonant masses; tar/sand damped vitreous structures; and so on.
So perhaps it's better to ask some questions than posit insubstantial hypotheses alongside assumptions as to another's technical acumen, particularly when said 'other' is one you probably do not even know.
Coming off looking as though you presume to teach your old grannie how to suck eggs will never put you in good stead.
I apologize if my tone comes across a little harsh, I've been in audio-related forums for 15 years, written probably thousands of posts and have by now seen a great deal of people's assumptions and very, very little of their simply asking questions.
Bill Perkins - PEARL
Last edited:
Yes, I've done the work. Vibrometers on boxes. Modal analysis. Spectral decay of box resonances.
30 years in the business and many products for KEF, McIntosh, JBL, Meridian, Snell, a/d/s and PSB.
Didn't realize I was stepping on any particular toes, I thought we were having a general discussion. The initial subject was whether it was worthwhile to vibration issolate woofers. I related what we had measured, heard and designed a fix for, at KEF.
If you are saying that you are the designer of the tweeter issolation mount, then perhaps you can explain the rational behind it.
David
Hi:
After much work I concluded that trying to decouple a bass unit from its enclosure was no part of a good idea and for at least two reasons:
- the decoupling compliance and the decoupled mass set up a 'Q-y' mechanical tank that sounds like hell on the best of days. The KEF and B & W implementations being classic examples,
- decoupled from the box or not the drive unit's acoustic output is more than sufficient to strike the high-Q box panel resonances into full essentially full 'song.'
Give the foregoing, my experience is that any attempted decoupling where the drive unit is not substantially mass loaded creates rather than solves problems; and even with the requisite mass loading the inherent box problems remain, unabated.
Bill Perkins - PEARL
After much work I concluded that trying to decouple a bass unit from its enclosure was no part of a good idea and for at least two reasons:
- the decoupling compliance and the decoupled mass set up a 'Q-y' mechanical tank that sounds like hell on the best of days. The KEF and B & W implementations being classic examples,
- decoupled from the box or not the drive unit's acoustic output is more than sufficient to strike the high-Q box panel resonances into full essentially full 'song.'
Give the foregoing, my experience is that any attempted decoupling where the drive unit is not substantially mass loaded creates rather than solves problems; and even with the requisite mass loading the inherent box problems remain, unabated.
Bill Perkins - PEARL
Toole has shown that hi-Q resonances are less audiable than lowQ ones, and further due to the narrower bandwidth under the curve, harder to excite.
dave
A couple of question about in-wall dampening.
I was thinking of using a 3 layer sandwich approach with "Green Glue". Would you know how this Noise Killer Yellow compares? A quick search found an interesting article on TNT-Audio regarding Noise Killer Yellow :
Noise Killer: Pro Damping technology
How does the dampening of MDF, BB plywood and a phenolic resin plastic (Paperstone) compare? I'm thinking of mounting the drivers to the middle layer in the snadwich and using the Paperstone for this layer. Combining a material that absorbs some of the vibration with a visco-elastic glue that absorbs some more seems like a good idea to me.
Well, this is not a simple matter.
A speaker enclosure has many, many resonant modes through at least the two octave span from 100 to 400Hz and Toole is right.
However, clusters of higher Q resonances will cross-excite (cross-entrain) to a great extent to produce what -appears- to be a lower Q, broader band mode but whose effect is of comprised of any number of individual mechanical 'tanks' whose respective outputs are less than coherent and whose summed output is lower than it could be if all points were in coherent motion.
A nasty problem . . .
BP - PEARL
A speaker enclosure has many, many resonant modes through at least the two octave span from 100 to 400Hz and Toole is right.
However, clusters of higher Q resonances will cross-excite (cross-entrain) to a great extent to produce what -appears- to be a lower Q, broader band mode but whose effect is of comprised of any number of individual mechanical 'tanks' whose respective outputs are less than coherent and whose summed output is lower than it could be if all points were in coherent motion.
A nasty problem . . .
BP - PEARL
My recollection was that the KEF and the B&W were both very highly regarded. "Sounds like hell" was not a quality I ever recall being ascribed to them. My memeory is getting a bit hazy these days, though.
Per our experiments at KEF, the decoupling dramatically reduced the drive to, and therefore the output from, the various box resonances. Yes, in the end the total improvement is limited by the fact that even without vibrational drive to the resonances there is acoustical drive. However this is after a very worthwhile 20dB or so reduction.
For Pit Hinder: External vibration applied to the tweeter is just another surface contributing to acoustic output. It causes its 1" diameter to drive the air. Wouldn't the same amplitude of vibration be considerably more detrimental if it were applied to the hundreds of square inches of any cabinet side?
In the end the issue is not so much cabinet vibration but the eveness of output and avoidance of excitation of high Q resonances that can linger considerably in time. Loudspeaker cabinets tend to have very high Q resonances and the usual attempts to correct that can move the resonances up in frequency but seldom reduce their effect.
Woofer decoupling is an effective treatment. I never heard why tweeter decoupling was useful?
David
Absolutely, this goes back to a paper by Peter Fryer. He found that every time you doubled Q the output had to be about 3dB louder for equal audibility. Now if we are talking a mechanical system and we double its Q, meaning we've cut frictional losses in half, output at resonance will go up by 6dB.
It becomes 3dB more discernable.
David
I can't comment directly on Green Glue but had good results with Noise Killer Yellow. In the end we laminated 2 layers of MDF, usually 2 x 1/2" and found it a practical solution. We could prelaminate it and run it through the NC router as if it was a standard material.
Read what you can about "constrained layer damping". The idea is that applying damping on a surface is inefective because the mechanical impedances are very high and a surface layer of goo has little impact. When you put the goo between two rigid layers you force the energy into a motion of shear and the goo is much more effective.
If you want to have drivers mounted to a central layer of a three layer sandwich then a softer coupling may work better at that (haven't tried it). In the end the noise killer/MDF layers sandwich is still fairly stiff, just better damped as a composite material.
David
David wrote:
"My recollection was that the KEF and the B&W were both very highly regarded. "Sounds like hell" was not a quality I ever recall being ascribed to them. My memory is getting a bit hazy these days, though.
Per our experiments at KEF, the decoupling dramatically reduced the drive to, and therefore the output from, the various box resonances. Yes, in the end the total improvement is limited by the fact that even without vibrational drive to the resonances there is acoustical drive. However this is after a very worthwhile 20dB or so reduction.
In the end the issue is not so much cabinet vibration but the evenness of output and avoidance of excitation of high Q resonances that can linger considerably in time. Loudspeaker cabinets tend to have very high Q resonances and the usual attempts to correct that can move the resonances up in frequency seldom reduce their effect.
Woofer decoupling is an effective treatment. I never heard why tweeter decoupling was useful?"
David
======================================================
Well, what can I say, I always hated the sound of decoupled bass drivers but then perhaps that's just me.
As to tweeter decoupling: I was messing around late one night and thought, "Hey, why don't you decouple the tweeter head assy, from the box itself and see what happens ? ?"
So I pulled out a couple of towels, folded them to the right size, put them under each tweeter head assy . . . and proceeded to amaze myself.
Note though that this was back in the days of rigid but only slightly damped bass unit enclosures. As those enclosures were greatly improved over the years the improvement from tweeter decoupling decreased.
BP - PEARL
"My recollection was that the KEF and the B&W were both very highly regarded. "Sounds like hell" was not a quality I ever recall being ascribed to them. My memory is getting a bit hazy these days, though.
Per our experiments at KEF, the decoupling dramatically reduced the drive to, and therefore the output from, the various box resonances. Yes, in the end the total improvement is limited by the fact that even without vibrational drive to the resonances there is acoustical drive. However this is after a very worthwhile 20dB or so reduction.
In the end the issue is not so much cabinet vibration but the evenness of output and avoidance of excitation of high Q resonances that can linger considerably in time. Loudspeaker cabinets tend to have very high Q resonances and the usual attempts to correct that can move the resonances up in frequency seldom reduce their effect.
Woofer decoupling is an effective treatment. I never heard why tweeter decoupling was useful?"
David
======================================================
Well, what can I say, I always hated the sound of decoupled bass drivers but then perhaps that's just me.
As to tweeter decoupling: I was messing around late one night and thought, "Hey, why don't you decouple the tweeter head assy, from the box itself and see what happens ? ?"
So I pulled out a couple of towels, folded them to the right size, put them under each tweeter head assy . . . and proceeded to amaze myself.
Note though that this was back in the days of rigid but only slightly damped bass unit enclosures. As those enclosures were greatly improved over the years the improvement from tweeter decoupling decreased.
BP - PEARL
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.